Abstract

This study addresses the issue of cloud parameterization in general circulation models utilizing a twofold approach. Four versions of the Florida State University (FSU) global spectral model (GSM) were used, including four different cloud parameterization schemes in order to construct ensemble forecasts of cloud covers. Next, a superensemble approach was used to combine these model forecasts based on their past performance. It was shown that it is possible to substantially reduce the 1–5-day forecast errors of phase and amplitude of the diurnal cycle of clouds from the use of a multimodel superensemble. Further, the statistical information generated in the construction of a superensemble was used to develop a unified cloud parameterization scheme for a single model. This new cloud scheme, when implemented in the FSU GSM, carried a higher forecast accuracy compared to those of the individual cloud schemes and their ensemble mean for the diurnal cycle of cloud cover up to day 5 of the forecasts. This results in a 5–10 W m⁻² improvement in the root-mean-square error to the upward longwave and shortwave flux at the top of the atmosphere, especially over deep convective regions. It is shown that while the multimodel superensemble is still the best product in forecasting the diurnal cycle of clouds, a unified cloud parameterization scheme, implemented in a single model, also provides higher forecast accuracy compared to the individual cloud models. Moreover, since this unified scheme is an integral part of the model, the forecast accuracy of the single model improves in terms of radiative fluxes and thus has greater impacts on weather and climate time scales. This new cloud scheme will be tested in real-time simulations.

Abstract

The main goal of this study is to investigate the relative contributions from the components of dynamics and physics of a forecast model, toward the understanding of the recurvature dynamics of hurricanes. A number of experiments were conducted using the Florida State University Global Spectral Model (FSU GSM), run at a global resolution of 126 waves. The method of physical initialization was used to “spin up” the model, 24 h prior to the 5-day forecast period to better define the initial water vapor, sensible heat fluxes, and rainfall rates.

The usage of the FSU GSM employed a partitioning of the dynamics and physics into separate components, that assumes a residue-free budget of the models' components. The model dynamics were broken down into a nonlinear advective component and also a linear dynamics (rest of the dynamics) partition. The model physics were partitioned into four components: deep convective heating, large-scale precipitation (nonconvective stable rain), total radiation, and shallow convection and surface fluxes. A total of four cases were examined, two each for Hurricanes Cindy and Dennis—1200 UTC 26 and 27 August, and 1200 UTC 28 and 29 August, occurring during the 1999 Atlantic hurricane season. The series of model runs were formulated to examine the tropical cyclone forecast tracks, suppressing one or more of the partitions for each time step, through day 5 of a forecast. Initial experiments coupling both the nonlinear advective and the linear dynamics (summed to equal the “total dynamics”) found that the total dynamics component resulted in a weakly recurving track for each of the storm cases. The addition of the physics components incrementally sharpened the recurving track through time.

While the full model dynamics was used as a baseline, the results of this study indicated that the deep convective heating (also referred to as deep convection) and total dynamics combined to produce a recurving track for both storms, for 50% of the four examined cases. The remaining cases required that the shallow convection and surface fluxes partition be included along with the deep convection and total dynamics. It was found that incremental improvements occurred with both the deep convective heating and shallow convection and surface fluxes partitions, however, the additions of the large-scale precipitation and radiation partitions did not significantly improve the forecast track relative to the full model, and their resulting magnitudes were significantly smaller than the rest.

Abstract

Results of computations of energy exchanges between waves and waves and between waves and zonal flows (for a tropical belt 15S to 15N at 200 mb for northern summer) are presented in this paper. These exchanges are 92-day averages. The procedure for calculations is essentially that proposed by Saltzman. The main results of this study are:

Zonal wavenumber 1 appears to be a major energy source for the tropics. In general, one of two other waves interacting with wavenumber 1 gains energy.

Wavenumber 1 also supplies energy to the zonal flows in the mean. Thus, it has a stabilizing influence on the zonal flows. On the other hand, zonal flows are unstable and lose energy to shorter waves, especially around wavenumber 8.

In this paper we also discuss some details of the nonlinear wave-wave energy exchanges.

Abstract

A global spectral model is used to carry out a number of short to medium range prediction experiments with global datasets. The primary objective of these studies is to examine the formation and motion of the hurricanes/typhoons with a fairly comprehensive state-of-the-art global model. Ale spectral model utilizes the usual transform method for the calculations of the nonlinear and physical processes. The physical processes include parameterizations of the planetary boundary layer, deep and shallow cumulus convection, radiative processes (including cloud feedback processes, diurnal change and surface energy balance) and large-scale condensation. ‘Envelope orography’ is used to represent steep mountains globally. Ocean temperatures are prescribed from a Preceding 10 day averaged dataset for the storm periods under investigation.

Sensitivity of storm forecasts to horizontal and vertical resolutions, datasets and representation of physical processes are addressed in this paper.

The major findings of this study are that improved results on the formation and motion of storms are achieved in several cases when (i) the surface layer fluxes are adequately resolved, (ii) the final FGGE analyzed datasets are used, (iii) very high resolution in the horizontal (106 waves triangular truncation) is used, and (iv) improved physical parameterization for the boundary layer, cumulus convection and radiative process are included.

The major limitation of this study is that in spite of the use of very high resolution the inner rain area (radius<150 km from the storm center) is not adequately represented to describe the central pressure, maximum wind and the warm core of hurricanes. Further studies to improve these areas are suggested.

Abstract

In the various formulations of Kuo (1965, 1974) type cumulus parameterization schemes, the moistening and heating by the cumulus are made proportional to the humidity and temperature differences between a model cloud and its environment. The constants of proportionality that differentiate the various versions of the Kuo's schemes are based on different closure assumptions. The proportion of available moisture supply that goes into the moistening by cumulus convection usually determines these constants. It is possible to diagnostically calculate the observed (or what might be called the exact) measures of these constants of proportionality. This enables one to define an ultimate Kuo scheme where the vertical integrals of the heating and moistening are exactly known but the vertical distributions are limited by the aforementioned structure functions. This ultimate Kuo scheme is not a prognostic scheme, but it serves as a benchmark in defining how far one can progress with this type of scheme in the prognostic sense. Using the final validated GATE B-scale data sets a comparison is made between the observed vertical distributions of the apparent heat source and apparent moisture sink (obtained from direct substitutions of observed data) with the ultimate Kuo scheme to assess its scope. Comparison of Kuo (1965, 1974) type schemes is next carried out with the ultimate Kuo scheme to address their limitations.

A proposal for a mesoscale convergence parameter η and a moistening parameter b is made to overcome some of the limitations of the above schemes. Here a multiple regression search of large-scale parameters, using 72 map times of data, is carried out to determine these parameters via least-square minimization of errors. These are next used to determine the vertical structure of moistening and heating, for a semi-prognostic formulation. The results show that by using the vertical average of the large-scale upward vertical motion and the lower tropospheric relative vorticity in the multiple regression, it is possible to attain an accuracy close to that prescribed by the ultimate Kuo scheme. Detailed results on the vertical distributions of the heating and moistening and the rainfall rates for the entire third phase of GATE are presented in this paper.

Abstract

The superensemble technique has previously been demonstrated to provide an improved seasonal forecast compared to the bias-removed ensemble of equally weighted models. This paper offers a further improvement to the superensemble method by modifying the regression coefficients used in the weighting of the models for the construction of the superensemble. The improvement is achieved by use of singular value decomposition of the covariance matrix, and selecting only the largest singular value, corresponding to maximal explained variance, for the calculation of the regression coefficients. The results shown here are based on calculations done with 10 yr worth of monthly forecasts from the Atmospheric Model Intercomparison Project (AMIP) dataset, using cross validation.

Abstract

This paper defines a mechanism for the genesis of tropical cyclones from African easterly waves (AEWs) over the eastern Atlantic, the so-called Cape Verde storms. Convective “superbursts” produce strong diabatic heating, which then strengthens the African easterly jet (AEJ), leading to enhanced barotropic energy conversions, which occur at the critical developmental stages of the system.

Diabatic heating is calculated using the Ertel isentropic potential vorticity (IPV) equation, while energy conversions are determined using energy equations first derived by Lorenz. The genesis mechanism is developed from studying Hurricane Bill (2009), as well as Tropical Storm Debby, Hurricane Helene, and a nondeveloping AEW, all from the 2006 NASA African Monsoon Multidisciplinary Analysis (NAMMA) field experiment, using the NCEP Final (FNL) analyses and the Advanced Research Weather Research and Forecasting model (WRF-ARW) simulations.

A striking and singular maximum in the diabatic heating due to the convective superburst is shown to precede by 24–36 h a pronounced maximum in positive barotropic energy conversion, which is demonstrated to occur simultaneously with the strengthening of the AEJ. The maximum in barotropic energy conversion is documented to occur in the developmental stages of the system, typically in the depression or early storm stages.

A physical mechanism is developed to explain how a mesoscale convective superburst can lead subsequently to an enhanced synoptic-scale AEJ over the eastern Atlantic, an enhanced jet that is critical to the genesis mechanism.

The findings agree with cited idealized studies by other investigators who found that moist AEWs grow 3 times stronger than dry waves as a result of faster AEJ development and larger barotropic energy conversions.

Abstract

The availability of daily observed rainfall estimates at a resolution of 0.5° × 0.5° latitude–longitude from a collection of over 2100 rain gauge sites over India provided the possibility for carrying out 5-day precipitation forecasts using a downscaling and a multimodel superensemble methodology. This paper addresses the forecast performances and regional distribution of predicted monsoon rains from the downscaling and from the addition of a multimodel superensemble. The extent of rainfall prediction improvements that arise above those of a current suite of operational models are discussed. The design of two algorithms one for downscaling and the other for the construction of multimodel superensembles are both based on the principle of least squares minimization of errors. That combination is shown to provide a robust forecast product through day 5 of the forecast for regional rains over the Indian monsoon region. The equitable threat scores from the downscaled superensemble over India well exceed those noted from the conventional superensemble and member models at current operational large-scale resolution.

Abstract

This paper addresses physical initialization of precipitation rates for a mesoscale numerical weather prediction model. This entails a slight modification of the vertical profile of the humidity variable that provides a close match between the satellite and model-based rain rates. This is based on the premise that the rain rate from a cumulus parameterization scheme such as the Arakawa–Schubert scheme is most sensitive to the vertical profiles of moist static stability. It is possible to adjust the vertical profile of moisture by a small linear perturbation by making it wetter (or drier) in the lower levels and the opposite at levels immediately above. This can provide a change in the moist static stability in order to achieve the desired rain rate. The procedure is invoked in a preforecast period between hours −24 and 0 following Krishnamurti et al. The present study is the authors’ first attempt to bring in this feature in a mesoscale model. They first noted that the procedure does indeed provide a much closer match between the satellite estimate of initial rain and that from the physical initialization for a mesoscale model. They have examined the impacts of this procedure for the initialization and short-range forecasts of a monsoon rainfall event and a hurricane. In both of these examples it became possible to improve the forecasts of rains compared with those from control runs that did not include the initialization of rains. Among these two examples, the results for the monsoon forecasts that deployed a uniform resolution of 25 km and the Grell and Devenyi scheme over the entire domain had the largest positive impact. The hurricane forecasts example also show improvement over the control run but with less impact, which may be due to heavy rains from explicit clouds in the nonhydrostatic model. Here the results did convey a strong positive impact from the use of the physical initialization; however, forecasts of very heavy rains carry smaller equitable threat scores. These require development of a more robust precipitation initialization procedure.

Abstract

The Florida State University Global Spectral Model forecast skill is calculated for several tropical systems in the Atlantic and Pacific Ocean basins and is compared to operational forecast skill. The series of forecasts were initialized using global analyses and supplemental satellite data. Track forecast errors were calculated for the storm series for control runs, enhanced runs, and operational forecasts made for the same time periods. Cumulative results for all modeled 1995 hurricanes are summarized.